Medical Imaging with Deep Learning

Medical Imaging with Deep Learning (MIDL 2025) is a focused conference dedicated to the application and theoretical development of deep learning methods for medical image analysis. The scope encompasses segmentation, detection, classification, registration, reconstruction, and synthesis across imaging modalities including MRI, CT, ultrasound, histopathology, and retinal imaging, with consistent attention to the particular challenges of working with medical data: small labeled datasets, domain shift, distribution shift across scanners, and the need for interpretable, uncertainty-aware predictions.

MIDL has established itself as the go-to venue for researchers who work specifically on deep learning for medical images rather than treating it as an afterthought in a broader CV or ML conference. Radiologists, pathologists, biomedical engineers, and machine learning researchers all contribute, making it a rare forum where clinical insight and algorithmic innovation meet on equal terms. Graduate students entering this field will find a rigorous, collaborative, and rapidly evolving community.